SpaceX #2: The Gang Solves the Gas Crisis
I can do this all month
CLICK HERE FOR INTRO - Background
CLICK HERE FOR PART 1 - NEPA Primer / FAA has no business permitting oil and gas facilities
CLICK HERE FOR PART 2 - Elon Musk’s Natural Gas Treatment Plant
CLICK HERE FOR PART 3 - SpaceX is building a pipeline and doesn’t feel the need to mention it
CLICK HERE FOR PART 4 - SpaceX: Drill baby, Drill
CHAPTER 2: IT’S A F*CKING GAS PLANT
Let’s, theoretically, say you’ve hyped up the development and release of a massive rocketship that will take humans to Mars.
Let’s also say that, for whatever reason, you’ve decided that neither hydrogen gas nor traditional rocket fuel (kerosene) will work for this rocket.
Natural gas seems like a great choice, doesn’t it? FERC pipeline natural gas is 97+ percent methane, a tiny bit of ethane, and is nearly devoid of contaminants. It’s also cheap and readily available.
Let’s say your rocket facility is less than 20 miles away from one of the biggest FERC spec natural gas pipelines in the country, which moves 2.6 billion standard cubic feet (scf) of gas every single day.
You call up the operator of the pipeline and say “yes, Mr. Pipeline Capacity Salesman, I’d like the ability to take 50-70 million scuffs of natural gas per day, but only on an as needed basis. Can you have your natural gas tech come and hook up the facility next week? That would be great.”
Off to Mars we go!
I bet you know where this is going, but for clarity: we don’t live in that world.
SpaceX’s facility is indeed mere miles away from the Valley Crossing Pipeline, owned by Midstream operator Enbridge. There’s a little problem though. You need natural gas that you can supercool and compress down into a liquid STAT, like 6 months from now. Enbridge’s pipe is currently at 100% of capacity. They don’t have 50 million scuffs for you. You need about 4-5 years after signing a capacity agreement with the pipeline to get your facility hooked up. Enbridge plays by the rules, too, so just getting the Pipeline to run underground through federally owned land, will require, ughhhhhh, following FERC requirements for planning and install of the pipeline modification. Plus Enbridge wants what’s called a “Take or Pay” contract, and they require a sizable down payment from you before they’ll even start doing anything.
Being the humongous genius that you are, you can see that gas is getting expensive anyways and you do, after all, run a company full of rocket scientists. So you assign some junior engineer to figure out how to get some gas without dealing with these stickler-for-the-rules pipeline companies.
Your crack team of engineers, who once took a college course on chemical process design, come up with a plan: it’s pure brilliance. There are some old oil and gas leases a few miles away from your facility. We’ll just take the gas out of the ground ourselves and then send it over to STARBASE, do some light “fuel pretreatment” using a supercooled separation system and then go on our merry way!
“This is a cryo gas plant” was my first thought when I read the section from the PEA submittal. I’ve spent enough time in oil and gas to know this, but never as a run-plant or design engineer. So to be safe, I sent the following screenshot to three engineers I know.
“Yeah that’s a gas plant, I used to run one”
“terminology is idiotic but that’s a f*cking cryo gas plant”
“Ya dude: gas plant… too many parallel processes are missing… no one calls them NG pretreatment systems. This was written by someone who has no idea how this stuff works”
Modern cryo gas plants can achieve up to 93% ethane removal and 98% of C3+ (propane and heavier) hydrocarbons. The outlet “residue” stream of one such cryo plant that feeds the Valley Crossing Pipeline looks like this:
Despite claims to the contrary, no gas coming out of such a plant will ever be “pure methane.” Assuring >99% purity is a service provided by companies like Air Liquide, who charge out the nose for certified methane cylinders used for laboratory equipment, and nothing you could get out of a combo 200 ft deethanizer column.
The stream above will have a nearly identical chemical profile to 100% methane for usage as a combustion fuel. A 1% ethane gas will not change compressability, condensation temperature or btu/scf except on an extreme margin and is functionally equivalent to Pure CH4.
SpaceX’s “pretreatment system” will remove “impurities such as water, carbon dioxide (CO2), and hydrocarbons heavier than methane.” You’ll notice that water is specifically mentioned as is CO2. The above processed gas stream is 0% water and 35 ppm CO2. This, of course, leads one to the realization that the input gas into the plant is probably important.
Here’s an example of what sweet field natural gas generally looks like in Texas:
This sample is an inlet stream from a plant that feeds into the Valley Crossing Pipeline (SpaceX’s neighbor). It’s really sweet stuff (low H2S), and may have been field treated prior to entry into the gas plant stream. Regardless, 26ppm is on the extreme conservative side for estimation. Certainly inappropriately optimistic for project evaluation purposes, but we’re giving SpaceX the benefit of the doubt here.
If that gas is treated, separated and all consumed on site, that little Sulfur molecule has to go somewhere, either as a SO2 molecule (fully combusted) or as deadly H2S from inadequate control.
Based on power plant sizing (1300 MMBtu/hr), and adding 20% for rocket fuel and utility/process heat, this facility demands at minimum 50 million scf wet gas per day. At 26ppm, and assuming all fuel is consumed on site (see next section as to why), that means that 24 tons of SO2/year can be assumed (based on full oxidation of 12.5 tons H2S)
1 TON! I’m sorry to focus on this, but it’s absolutely insane.
As an aside, I don’t think SpaceX knows anything about the nature of the field gas coming into the facility, and I suspect they haven’t done any analytics on it. At a very possible >100 ppm H2S, that SO2 per year goes up to 100 tons. This Emissions inventory is wildly incomplete to an almost incredible degree, and indicative of the lazy and Non-NEPA compliant assessment released by FAA.
Lots of missing information
“AAAHHHHH YOU MADE BAD ASSUMPTION” is what some SpaceX super fan will screech at me any time I try to fill in the blanks. Here’s the problem:
NEPA requires full disclosure of all operations and a full inventory of impacts, and it’s not my job to assure this is the case
When I make assumptions, I always try to err on the side of conservatism (eg assuming all field gas is extremely sweet), so usually your “gotcha” will just make more issues
I calculated fully combusted H2S on site, which pushed up the SO2 emissions, yes. But Why?
Well, if that sulfur comes into the plant, it has to go out somewhere. The concentration of H2S into the residue (97% methane) will be minimal. It might be part of a overhead waste stream that is fed into a process heater at the gas plant (which would be in the facility SO2 emissions). Or it might go into a Y-grade (propane/butane) fraction or a condensate fraction.
Here’s the problem. SpaceX doesn’t acknowledge these fractions exist.
Y-grade gasses typically go out by either pipeline or truck. We know there’s not a Y-grade pipeline (don’t trust me? check PHMSA). Y-grade storage would require one of these things:
There’s no mention of any Y-grade storage on the PEA (odd huh?). If they had one of these spheres, the facility SO2 emissions using absolute mass balance might go down, but then the VOC emissions would skyrocket (loading and storage).
Natural Gas Liquids - these heavier liquids, sometimes referred to as condensate, are a pungent, gasoline like liquid. Now I know for 100% certain that SpaceX’s interns didn’t assume that these are burned on site. You know why?
The Particulate Emissions (which are laughably small as is) would be significantly higher. Liquids make more smoke than gas when burned
That means that condensate NGLs are being stored on site and must be trucked off. Again, this activity is not mentioned in the PEA: another egrigious violation of NEPA, etc etc
And finally, speaking of smoke:
There’s no mention of a process flare, or a thermal oxidizer. Both of which would be required per the TCEQ Oil and Gas standard permit. You know, one of these things:
Also missing is a description (or even existence) of the following critical pieces of equipment:
Oil Storage Tanks (size or location) ✔️- NEPA VIOLATION
Amine oil heater (no size, heat input or waste streams) ✔️- NEPA VIOLATION
Thermal Oxidizer (Combustion Control) ✔️- NEPA VIOLATION
Methanol storage tanks (size + waste streams) ✔️- NEPA VIOLATION
Produced water storage and disposal ✔️- NEPA VIOLATION
Process Flare ✔️- NEPA VIOLATION
Condensate loading equipment ✔️- NEPA VIOLATION
By the way: this is all stuff from the top of my head. The omissions are truly striking, folks.
More to come, ESGH
Almost everyone who sweetens gas streams uses iron oxide columns and/or triazine scavenger in contact towers to control H2S. Where does this fit into your H2S / SO2 analysis?